The stress-strain relationship curve is the most intuitive method for evaluating soil deformation and strength characteristics. Figure 11 shows the stress-strain curve of PCM-modified expansive soil, where N represents the number of F-T cycles. In this test, the stress-strain curves all show a strain-softening type. Taking the pure soil sample as an example, the entire stress-strain curve can be divided into four stages. In the first stage, because the uneven surfaces at both ends of the sample cannot make good contact with the loading end, it cannot reflect the true mechanical properties of the sample. In the second stage, as the strain increases, the stress rises approximately linearly. As the number of F-T cycles increases, the rising rate slows down, which is regarded as the elastic stage. In the third stage, the stress rise rate gradually slows down, and then the stress peak appears. Due to the gelation between the expanded clay particles, the sample undergoes obvious and uniform plastic deformation, but the soil is damaged by the F-T cycle. The duration of plastic deformation is significantly reduced, and it stabilizes after 3 times, which is regarded as the damage stage. In the fourth stage, the stress gradually decreases as the strain increases, which is regarded as the brittle fracture stage. From Fig. 11(b), (c), (d), it can be found that the changing trend of the stress-strain curve of PCM-modified expansive soil is quite different under different content. With the increase of PCM content, the approximate linear increase of stress in the elastic phase becomes more obvious. The plastic characteristics are weakened in the damage stage; the greater the amount, the more obvious the weakening degree. In the brittle fracture stage, the rate of stress decrease increases. When the content is 10%, it decreases linearly until the strength is completely lost.
4.2.2 Failure strain
Failure strain is an important index to measure the deformation characteristics of the soil. For general strain-softening soils, the strain corresponding to the maximum stress value is considered the failure strain (Gao et al. 2019). Figure 12 reflects the relationship between the failure strain of the sample and the number of F-T cycles. It can be seen from Fig. 12 that the first F-T cycle has a greater impact on the failure strain of the pure soil sample, and then tends to be stable. For the expansive soil treated by PCM, the failure strain is less affected by F-T cycles and remains unchanged. This shows that the PCM improved expansive soil is stable in a certain sense. Under various PCM contents, the failure strain value of the sample is lower than that of pure soil. The overall trend of failure strain is decreasing with the increase of PCM content. The above phenomenon can be explained as PCM shell is a polymer material (Ashraf 2019), which has a certain degree of brittleness, and the compressive strength of PCM is small, thus aggravating the brittleness of soil.
4.2.3 Effect of freeze-thaw cycles on soil mechanical strength
For general strain-softening soils, the stress corresponding to the peak of the stress-strain curve is taken as the UCS (Cao et al.2019). The variation of the unconfined compressive strength of the sample with the number of F-T cycles is shown in Fig. 13. It can be seen from Fig. 13 that the first F-T cycle has the most significant attenuation effect on the strength of the soil, which can account for more than 53.4% of the entire F-T cycle test. This is because under the dual influence of the complex ice-water phase change and the expansion and shrinkage of soil grains, the soil structure is severely damaged, and the pore distribution changes. With the increase in the number of F-T cycles, the soil strength's attenuation gradually slowed down and stabilized after the third time. PCM can inhibit the attenuation of soil strength due to freezing and thawing cycles. The greater the PCM content, the better the inhibition effect. After 7 cycles, the strength attenuation rates of the samples with PCM content of 5%, 8%, and 10% were 60.7%, 46.7%, and 37.7%, respectively, and the unconfined compressive strengths were all higher than those of pure soil samples. This is because PCM can store or release heat in the form of latent heat during the F-T cycle to resist damage to the soil caused by external temperature changes, thus slowing soil strength attenuation.
Regression analysis was carried out on the unconfined compressive strength test results of PCM-modified expansive soil under different PCM content and different F-T cycles. A mathematical model between the unconfined compressive strength and the number of F-T cycles is established, which can be described by an exponential equation, and the fitting curve is shown in Fig. 13. The specific fitting function is shown in the following formula: . In the formula:
q u is the unconfined compressive strength;
A, B, and C are the coefficients related to the PCM content, whose values are shown in Table 4;
N FT is the number of F-T cycles.
Combining Table 4 and Fig. 13 shows that under the same PCM content, the unconfined compressive strength and the natural index of the number of F-T cycles have an obvious linear negative correlation. The coefficient A + B is the theoretical value of the unconfined compressive strength in the initial state (0 cycles) under a certain PCM content. The coefficient C reflects the rate at which the unconfined compressive strength at a certain PCM content decreases with the increase in the number of F-T cycles. The absolute value of the coefficient C shows a decreasing trend with the increase of PCM content, which shows that the greater the PCM content affected by the effects of F-T cycles, the slower the unconfined compressive strength of the sample decays. In the three sets of data, R2 is close to 1, indicating that the fitting effect is better and can more truly reflect the functional relationship between the two.
Table 4 Function fitting results.
4.3 Thermal stability
The expansive soil sample in a dry state is tested, and it is found that the dehydrated soil particles do not undergo energy conversion during the DSC test. Therefore, it can be considered that only water undergoes a phase change. The DSC curves of expansive soil and PCM-modified expansive soil are shown in Fig. 14. Due to the enormous energy released during freezing, the temperature of the sample rises again so that the DSC curve will be bent back. In Fig. 14, the upward peak represents the endothermic curve peak, the downward peak represents the exothermic peak, and the curve peak area represents the phase change latent heat. Because the DSC curves of different PCM content are similar in shape, limited to space, only the DSC curves with 0 and 10% PCM content are listed. It can be seen from Fig. 7 that the incorporation of PCM reduces the latent heat of phase change of the expansive soil by about 10.93%, which means that the heat released by the phase change of PCM at 4.61 ~ 1.63℃ is "stored" in the soil particles, which may cause some water did not phase into ice. After adding PCM, the starting point of the "primitive peak" phase transition temperature changed from − 7.61°C to -6.30°C, which indicates that the soil body mixed with PCM has increased in temperature during freezing and melting. At the same time, it can be found that the width of the phase transition peak has also increased slightly, and the supercooling phenomenon has also been slowed down. This indicates that the incorporation of PCM delays the formation of ice lenses in the test temperature range, which helps improve the internal temperature field of the canal soil. It is beneficial to improve the thermal stability of the soil.
Table 5
thermal properties of samples.
Property
|
pure soil
|
10% PCM-modified soil
|
Melting
|
|
Temperature Range
|
-7.61 ℃ to -10.83 ℃
|
4.61 ℃ to 1.63 ℃
|
-6.30 ℃ to -10.71 ℃
|
Peak Temperature
|
-7.74 ℃
|
2.58℃
|
-6.34℃
|
Enthalpy
|
60.82J/g
|
19.34J/g
|
54.17J/g
|
Crystallization
|
|
Temperature Range
|
-2.18 ℃ to 2.42 ℃
|
4.73℃ to 7.55 ℃
|
-1.24 ℃ to 2.62 ℃
|
Peak Temperature
|
0.93 ℃
|
6.15 ℃
|
0.97 ℃
|
Enthalpy
|
61.74J/g
|
18.78J/g
|
54.36J/g
|
4.4 Micro mechanism
Figure 15 are 100X scanning electron microscope images of the sample with pure soil and 8% PCM content, where black represents pores or cracks and white represents soil. It can be seen that in the initial state (0 cycles), the soil particles are cemented and connected to form a whole, and the soil sample shows good integrity. After the sample undergoes a F-T cycle, the pores of the sample begin to develop. The pores after 7 F-T cycles were significantly larger than the pores after 1 F-T cycle, and obvious through-fractures formed inside the soil after 7 F-T cycles. The incorporation of PCM allows the large pores of the soil to be filled, and the many fine pores formed to make the pore canals narrow and tortuous, reducing the connectivity of the pores, hindering the flow of water, and inhibiting the expansion and contraction characteristics of the expansive soil. Figure 16 is obtained after local magnification of the PCM-modified soil sample by 1000 and 5000 times. It can be seen that the low content of PCM is mainly distributed in the pores of the soil. As the content increases, PCM begins to aggregate into a granular structure, which is one of the important reasons for the slight decrease in the mechanical strength of the soil in the initial state.
Based on the professional image processing software PCAS, quantitative analysis of the image scanned by the electron microscope can be used to extract parameters such as soil shape coefficient, pore direction and size. In this paper, we select an appropriate threshold to binarize the image under 100 times, and reduce the noise to segment the pores, and obtain the surface porosity of the sample after different F-T times (the proportion of pores on a certain plane of the soil), such as Shown in Fig. 17. With the increase in the number of F-T cycles, the surface porosity of the soil sample gradually increased, and the increase gradually slowed down after 3 cycles. After the end of the 7 cycles, the porosity of the soil samples with the four PCM content increased by 15.40%, 12.78%, 9.29% and 6.01%. Macroscopically, PCM weakens the effect of F-T cycles on the degradation of soil mechanical properties; that is, the greater the PCM content, the slower the unconfined compressive strength attenuation. At the same time, it can be found that in the early stage of the test, the lower the PCM content, the faster the soil porosity increases. When the PCM content is low, the heat released is not enough to resist the liquid water phase changing into ice crystals and the damage to the surrounding soil particles, the more obvious the changes in the microstructure of the soil after the F-T cycle.